nai

Project Progress

We have studied the mineralogy and microscopic textures of four banded iron formations (BIFs): Isua, Greenland (3.8 Ga); Hamersley, Western Australia (2.5 Ga); Transvaal, South Africa (2.5 Ga), and Biwabik, Minnesota, USA (1.9 Ga). These rocks have been interpreted to preserve records of conditions and chemistry of Precambrian oceans. This hypothesis is evaluated with in situ SIMS analysis of oxygen (δ¹⁸O) and silicon isotope ratios (δ³⁰Si) in chert domains within the BIFs. The paragenesis of magnetite and hematite is important to understanding the fluid and thermal history of unmetamorphosed or low-temperature (sub-greenschist facies) oxide-facies BIFs. We identified silician (up to 3 wt% SiO₂) magnetite overgrowths in samples of the Dales Gorge BIF from Wittenoom. We hypothesize that silician magnetite is stabilized by organic matter and thus is a biosignature. We investigated the distribution of δ¹⁸O values in quartz, magnetite, silician magnetite, and hematite by SIMS and found that while quartz is homogeneous (average δ¹⁸O = 22.0 ±1.3‰ 2SD, VSMOW), values of δ¹⁸O for magnetite and hematite vary by 13‰ (-5 to +7‰) and are correlated with different generations of magnetite and hematite.

Correlated δ¹⁸O and δ³⁰Si analyses in BIF quartz show that while δ¹⁸O values are affected by metamorphism, δ³⁰Si values display a similar range of 2-3‰ for all BIFs studied regardless of metamorphic grade or age. Thus δ³⁰Si values are relatively insensitive to metamorphism and may retain primary depositional values. The δ³⁰Si values are homogeneous to ±0.6‰ within individual sub-mm microlaminae, and during metamorphism exchange of Si is restricted to the sub-mm scale (Fig. 1). Further, high δ³⁰Si values of up to +1.3‰ for Isua BIF suggest that continental weathering was locally important as early as 3.8 Ga.

Variation in δ¹⁸O values between quartz and magnetite are commonly used to assess equilibrium isotope fractionation in banded iron formations. Previous reports found that the δ¹⁸O values of quartz from the Dales Gorge BIF were either homogeneous at 22.0 ±1.3‰ (2SD, VSMOW, Becker and Clayton 1976, bulk analysis) while a more recent study found the δ¹⁸O values for quartz varied by 13‰ (11 to 24‰, Hayashi et al. 2008 and Ohmoto 2006, laser probe). In our study, the δ¹⁸O analyses of magnetite, silician magnetite, and hematite are precise to ±1‰ (2SD) employing modified SIMS analytical conditions developed to minimize variability in measured δ¹⁸O analyses due to crystal orientation effects (Kita et al., 2010; Huberty et al., 2010b). We determine a paragenetic sequence for iron oxides based on their δ¹⁸O values by SIMS and textural characteristics: magnetite sub-microlaminae, recrystallized magnetite fragments, silician magnetite overgrowths, euhedral low-Si magnetite euhedra, and anhedral hematite laminae (Huberty et al., 2010c). Figure 2 shows a plot of δ¹⁸O(magnetite) vs. δ¹⁸O(quartz) for BIFs worldwide. The δ¹⁸O zoning in magnetite microlaminae from the Dales Gorge Member reflects primary depositional values while δ¹⁸O values for quartz were homogenized during diagenesis and low temperature metamorphism (<300°C).

We used the UW-Madison CAMECA ims-1280 ion microprobe to develop a silicon isotope standard for quartz. The spot-to-spot precision in δ³⁰Si is ±0.2‰ (2SD, n = 14) for our quartz standard UWQ-1, which is more homogeneous than grains of NBS-28 quartz (±0.3, 2SD, n = 26).

In samples of quartz from the four BIFs, values of δ³⁰Si show a range of ~5‰, from -3.7‰ to +1.2‰ and extend to the lowest values reported for the Precambrian. The average δ³⁰Si for Isua is -1.6‰ while for Hamersley, Transvaal, and Biwabik, the average δ³⁰Si value is identical and 1‰ higher at -0.6‰ (Heck et al., 2011). The range in δ³⁰Si values is 4.5‰ for Isua, 3.5‰ for Hamersley, 2.7‰ for Biwabik, and 2.6‰ for Transvaal. Values of δ¹⁸O range from +7.9‰ to +31.4‰, including the highest reported δ¹⁸O values in BIF quartz (Huberty et al., 2010c; Heck et al., 2011). We interpret the narrow range in δ¹⁸O for Transvaal and Hamersley BIFs to reflect homogenization of quartz, thus these samples are not in equilibrium with seawater and cannot be used for oxygen isotope thermometry.

Individual amphibolite-facies samples from Isua are homogeneous in δ¹⁸O at the cm-scale, but show similar variability in δ³⁰Si as BIFs that have not experienced high temperature metamorphism. The wide range of δ³⁰Si at Isua suggests that silicon isotopes are not homogenized at the cm-scale by amphibolite facies metamorphism, as is observed for δ¹⁸O values. Most of our δ³⁰Si values fall below the mantle value of -0.4‰ and agree with the range of hydrothermal silica (δ³⁰Si = -0.4 to -3.0‰), while δ³⁰Si values greater than -0.4‰ may suggest a continental source. Thus our data show a mixture of both sources for Si in BIF quartz.

We have studied in detail the mineralogy of magnetite from the Dales Gorge Member of the Brockman Iron Formation, Hamersley Group, Western Australia. Figure 3 shows a BSE-SEM image where contrast has been adjusted to show only magnetite (Huberty et al., 2010a, 2011). Quartz, carbonate, and stilpnomelane are present but appear black in this image. Internal magnetite sub-microlaminae (white/ light grey) contain numerous μm to sub-μm inclusions of quartz, carbonate, apatite, and stilpnomelane whereas silician magnetite overgrowths (darker grey) are devoid of mineral inclusions. In situ micro-X-ray diffraction shows that the lattice parameter for silician magnetite is 8.388(2) to 8.392(2) Å (Huberty et al., 2011), lower than the reference value for pure magnetite, 8.397 Å, indicating that silicon is present in magnetite as solid solution. Similarly, high resolution bright field TEM images and select area diffraction patterns confirm the absence of quartz micro-inclusions and that silicon is in solid solution in magnetite. Silician magnetite is the dominant Fe-oxide in the Dales Gorge BIF. We hypothesize that the former presence of organic matter may be required to attain the low oxygen fugacity necessary to stabilize silicon in magnetite and propose that silician magnetite is a biosignature.